Repairing inconsistent answer set programs using rules of thumb: A gene regulatory networks case study

Abstract

Answer set programming is a form of declarative programming that can be used to elegantly model various systems. When the available knowledge about these systems is imperfect, however, the resulting programs can be inconsistent. In such cases, it is of interest to find plausible repairs, i.e. plausible modifications to the original program that ensure the existence of at least one answer set. Although several approaches to this end have already been proposed, most of them merely find a repair which is in some sense minimal. In many applications, however, expert knowledge is available which could allow us to identify better repairs. In particular, we consider the scenario where this expert knowledge is formulated as rules of thumb, but no training data is available to learn how these rules of thumb interact. The main question we address in this paper is whether we can then still aggregate the rules of thumb in a useful way. In addition to standard aggregation techniques, we present a novel statistical approach that assigns weights to these rules of thumb, by sampling, in a particular way, from a pool of possible repairs. In particular, we evaluate how frequently each given rule of thumb is violated in the sample of repairs, and use the Z-score of this distribution to set the weight of that rule. We analyze the potential of using expert knowledge in this way, by focusing on a specific case study: Gene Regulatory Networks. We describe the rules of thumb that express available expert knowledge from the biological literature and explain how they can be encoded while repairing inconsistencies. Finally, we experimentally compare the proposed repair strategies using rules of thumb against the baseline strategy of identifying minimal repairs.

Semi-automatic model revision of Boolean regulatory networks: confronting time-series observations with (a)synchronous dynamics

“…Furthermore, the proposed tool currently produces all the optimal solutions given the defined optimisation criteria. Heuristics to filter the produced output could also be considered, in particular those proposed in the work of Merhej et al (2017) where rules of thumb are discussed.…”

“…Most approaches define atomic repair operations and minimise the number of operations applied to render a model consistent. Merhej et al (2017) proposed the use of rules of thumbs, which are properties found in the literature, in order to repair inconsistent Boolean models. This work considers Boolean models where the regulatory function of each compound is as follows: it becomes active if there is at least one active activator and there is no active inhibitor; it becomes inactive if there is at least one active inhibitor and there is no active activator; otherwise its value is not changed.…”

“…In this work, we propose a new model revision approach of Boolean models that considers time-series observations and the network' (a)synchronous dynamics, addressing some of the limitations of previous approaches (Section 3). In particular, we consider monotone nondegenerate Boolean functions as regulators, allowing for more expressive regulatory functions than some of the previous approaches (Siegel et al, 2006;Gebser et al, 2010;Guerra and Lynce, 2012;Merhej et al, 2017). We define an optimisation criteria for possible model repairs which takes into account how these models are built, and the different degrees of confidence in the correctness of model components (interactions and regulatory functions).…”

“…To represent a Boolean logical model, we consider the Answer Set Programming (ASP) language due to its expressive power. Moreover, ASP has been used to represent biological models and proved useful in model inference, consistency check, and model revision approaches (Siegel et al, 2006;Gebser et al, 2010;Merhej et al, 2017).…”

“…The notion of belief revision exists for a long time and addresses the problem of having new information in conflict with previously known information (Alchourrón et al, 1985). Even though model revision has been addressed by the community for models of regulatory networks (Gebser et al, 2010;Guerra and Lynce, 2012;Merhej et al, 2017), few approaches have been proposed for (Boolean) logical models (Mobilia et al, 2015;Lemos et al, 2019).…”